Materials Science & Engineering A 688 (2017) 416–428

Contents lists available at ScienceDirect

Materials Science & Engineering A

journal homepage: www.elsevier.com/locate/msea

Effects of niobium addition on microstructure and tensile behavior of as- MARK

cast ductile iron
⁎
Xiangru Chena, Jie Xua, Henry Hub, , Hardy Mohrbacherc, Ming Kangd, Wei Zhange,
⁎
Aimin Guoe, Qijie Zhaia,
a
State Key Laboratory of Advanced Special Steel & Shanghai Key Laboratory of Advanced Ferrometallurgy & School of Materials Science and Engineering,
Shanghai University, Shanghai 200072, China
b
Department of Mechanical, Automotive and Materials Engineering University of Windsor, 401 Sunset Avenue, Windsor, Ontario, Canada N9B 3P4
c
NiobelCon bvba, Swaenebeecklaan, 2970 Schilde, Belgium
d
Dongfeng Commercial Vehicle Co., Ltd., Wuhan 430056, China
e
CITIC Metal Co., Ltd., Beijing 100004, China

A R T I C L E I N F O A BS T RAC T

Keywords: The effects of niobium addition up to 0.11 wt% on the microstructure and tensile properties of as-cast ductile
As-cast ductile iron iron (ACDI) were investigated. Metallographic analyses by both optical microscopy (OM) and scanning electron
Niobium microscopy (SEM) indicated that niobium (Nb) promoted the formation of pearlite, reduced pearlite lamellar
Microstructure spacing and decreased the extent of graphitization taking place in the Nb-alloyed ACDI. The nodularity and
(Nb, Ti) C nano precipitate
nodule counts of graphite changed insignificantly when the Nb content was less than 0.08 wt% in the ACDI. The
Tensile behavior
analysis of precipitates by transmission electron microscopy (TEM) revealed that nano and micro sized (Nb, Ti)
Fracture
C carbides acted as nucleation site for graphites, and promoted the formation of large graphite nodules with low
roundnesses as Nb content rose above 0.08 wt%. The results of tensile testing showed that the yield strength,
ultimate tensile strength and elongation of the ACDI with 0.08 wt% Nb increased by 12.1%, 11.2% and 14.3%
over those of the Nb-free ACDI, respectively. The optimum values of the yield strength, tensile strength and
elongation of the Nb-alloyed ACDI were found to be 418 MPa, 746.0 MPa and 8.0%, respectively, at the Nb
content of 0.08 wt%. The high strain hardening rates of the Nb-containing ACDIs implied that they were capable
of spontaneously strengthening itself increasingly to a large extent, in response to a slight plastic deformation
after yielding.

1. Introduction
More than 90% by weight of metallic materials used by human
beings are ferrous alloys, which are classified into two groups based on
the carbon (C) content in the alloys. Steel generally contains between
0.04 and 1.7 wt% C, while cast irons have between 1.8 and 4.0 wt% C
[1]. Compared to steel, cast irons have relatively low melting tempera-
tures, very good fluidity and castability, and moderate shrinkage during
solidification and cooling [2,3]. But, engineering applications of cast
irons were limited until ductile irons were successfully developed by
adding a small amount of magnesium and/or cerium to the gray iron in
1948 [4,5]. Nowadays, ductile irons are widely used for manufacturing
wheels, bearings and gears due to its excellent castability, good
mechanical properties and low cost [3–8]. Due to the spherical
graphite morphology, the strength of ductile iron is equivalent to that
of carbon steel. Iacoviello et al. [9] found that graphite nodules present
in the matrix of ductile iron behaved almost like voids with regard to
mechanical properties. Ductile iron comprising graphite nodules with
high counts and low average diameters exhibited better mechanical
properties. In the past decades, the implementation of heat treatment
processes and the introduction of alloying elements have been proven
to be critical for the mechanical property improvement of ductile iron.
Resulting from the application of heat treatment, austempered ductile
iron (ADI) is one of the most studied variants for its excellent tensile
strength, ductility and wear resistance [10–12]. However, increased
manufacturing cost and extended production cycle limit a widespread
application of austempered ductile iron. Research interests [13–18]
have also been in developing alloying methods, in which the effects of
alloying elements, such as copper, nickel or molybdenum on the
improvement of the tensile strength, hardness, wear resistance and

⁎
Corresponding authors.
E-mail addresses: cxr16@shu.edu.cn (X. Chen), shuxujie@163.com (J. Xu), huh@uwindsor.ca (H. Hu), hm@niobelcon.net (H. Mohrbacher), kangming@dfcv.com.cn (M. Kang),
zhangwei3@citic.com (W. Zhang), guoam@citic.com (A. Guo), qjzhai@shu.edu.cn (Q. Zhai).

http://dx.doi.org/10.1016/j.msea.2017.01.032
Received 17 July 2016; Received in revised form 29 December 2016; Accepted 9 January 2017
Available online 30 January 2017
0921-5093/ © 2017 Elsevier B.V. All rights reserved.
X. Chen et al.
ductility of ductile iron are investigated. The introduction of different
alloying elements into ductile iron enables the control and manipula-
tion of phase constitution, microstructural refinement and solution
strengthening. Consequently, the mechanical properties of ductile iron
are improved and even optimized to achieve both high strengths and
ductilities. Among various alloying elements, niobium is widely used
for microstructural control in nickel-base alloys [19–21], steel [22–24]
and cast irons [25–33]. Smith and Patel [21] pointed out that niobium,
as a refractory element, was capable of improving mechanical proper-
ties of nickel-base alloys through carbide formation and precipitation
hardening despite its limited benefit as a solid solution strengthener
due to its relatively low melting point and modulus compared to other
refractory elements (Mo. Ta and W). The review given by Morrison [22]
recognized the advantages of adding a small amount ( < 0.12 wt%) of
niobium to C–Mn steels for property improvement. The Nb addition
not only refined grain structure but also precipitated nano-sized NbC or
NbCN particles during normalizing treatment. Mohrbacher [23] ob-
served similar phenomena in multiphase steels microalloyed by Nb ( <
0.2 wt%).
Nb microalloying to grey cast iron and its influence on mechanical
properties are widely studied [25–29]. Zhou et al. [25] found that,
adding about 1.48 wt% Nb to grey cast irons increased hardness and
wear resistance considerably due to the massive presence of large-sized
NbC phases with dimensions over 10 µm, an increase in the number of
eutectic cells, and a reduction in the graphite size and the pearlite
interlamellar spacing. The study by Devecili and Yakut [27] showed
that the addition of 0.65 wt% Nb led to the formation of 10 µm chunky
Nb and Ti containing phases in grey cast irons, and improved their
abrasion resistance and tensile strength. It has been indicated [26]
that, with niobium level above 0.1 wt% primary NbC carbides could be
formed in the liquid iron, which acted as nuclei for the eutectic
reaction, i.e., the transformation of the liquid phase into austenite
and graphite or cementite. This explained the observation, that the
eutectic cells became finer with niobium additions. When the niobium
content exceeded the alloying limit of 0.2 wt%, such primary carbides
were formed already at elevated temperature becoming coarser with
time and could thus be found in the microstructure as NbC even by
optical microscopy. Zhu et al. [29] introduced 0.11 wt% Nb into grey
cast irons and observed their improved high-temperature tensile
strength and oxidation resistance. It was noted that brittle breaking
occurred with the specimen containing niobium of 0.037 wt%, and
ductile breaking occurred with the specimen containing niobium of
0.11 wt%, and the fracture exhibited ductile dimples. Fraś et al. [30]
studied the effect of small additions of niobium (up to 0.038 wt% Nb)
on structure and mechanical properties of ductile iron. It was indicated
that niobium increased graphite nodule count with small diameters
and the fraction of carbides in ductile iron, although the effect of small
additions of niobium on type of matrix is negligible. The Nb-alloyed
ductile irons exhibited the improved tensile and yield strengths, but the
reduced elongation. The work by Souza et al. [31] showed that the
0.47 wt% Nb addition to nodular cast irons (NCIs) led to a 20%
increase in their yield and tensile strengths over the values for the Nb-
free NCI because of the increase in pearlite content. A modest
increment of tensile ductility and Charpy toughness appeared while
only 0.23 wt% Nb was introduced. Alias et al. [32] alloyed the ductile
iron with 0.5–2.0 wt%. Their results showed that the Nb addition in
ductile iron provided significant enhancement in mechanical properties
when compared to unalloyed ductile iron. Addition of higher amount of
niobium had further increased the strength and impact toughness
properties. Bedolla-Jacuinde et al. [33] investigated the effect of
niobium in the range of 0–0.8 wt% in Ni-containing ductile cast irons.
Niobium was observed to be directly related to the formation of
polygonal niobium carbides of the type NbC, which had the size of as
large as 8 µm in the irons with 0.8 wt% Nb. The amounts of pearlite
and ferrite phases were not affected by the niobium content added in
their study. Neither nodule count nor nodularity was affected by such
417
Materials Science & Engineering A 688 (2017) 416–428
niobium addition. The Nb effect on the microstructure of the Ni-
containing ductile iron contributed to a small increase in yield and
tensile strength as well as in hardness along with a decrease in ductility.
So far, it appears that Nb addition to ductile iron has not been
extensively studied, in particular at a microalloying level of up to
0.10 wt%, although, in principal, it could provide microstructural
refinement and precipitation strengthening due to its strong carbide
forming tendency. It is of interest whether similar positive effects of Nb
microalloying on the mechanical properties of ductile cast iron could be
obtained.
In the present work, the effects of Nb microalloying up to 0.11 wt%
on the microstructure of the as-cast ductile iron (ACDI) were studied
using metallographic analyses. The microstructural analyses by optical
microscopy (OM), scanning electron microscopy (SEM) and transmis-
sion electron microscopy (TEM) revealed that phase constituents and
constitution as well as phase morphology in the ACDI were influenced
by Nb addition. The mechanical properties of the Nb-alloyed ACDI
were evaluated by tensile testing. Based on the results of tensile testing,
the optimum content of Nb with respect to both improved tensile
strengths and elongations was identified.
2. Experimental procedure
2.1. Material preparation
The raw materials employed in this study were pig iron, steel scrap,
80MnFe, copper, nickel and ferro-niobium (Fe65Nb). All the materials
were melted and homogenized at 1520–1540 ℃ in an induction
furnace having a maximum capacity of 16 kg. Upon the completion
of placing 1.2 wt% Fe-Si-Mg-Ba alloy (41.3 wt% Si, 6.9 wt% Mg and
1.52 wt% Ba) on the bottom of the preheated ladle (covered with an
iron sheet), the melt was poured into the ladle at a temperature of
1440–1460 ℃ for spherodization. After skimming the slag off the
surface of the spherodized melt, 0.8 wt% Fe-Si-Al alloy (74–79 wt%
Si, 0.8–1.6 wt% Al and 0.5–1.0 wt% Ca) was added for inoculation. At
last, all the melts were poured into a Y-block sand mold as illustrated in
Fig. 1. It should be noted that a small amount of melt was poured into a
metal mold for producing white iron specimens with a dimension of
0.004×Φ0.040 mm, which were used for the chemical composition
verification. The results of chemical analyses are shown in Table 1.
2.2. Microstructural analysis
Metallographic specimens were sectioned, mounted, and polished
from the bottom of Y-type specimens, and prepared following the
standard metallographic procedure. After polishing and 4% Nital
etching of metallographic specimens obtained from the bottom of Y-
Fig. 1. Dimensions of the Y-block sand casting employed in the present study, and the
arrow-pointed location in which specimens were sectioned for metallographic analyses
and mechanical testing.
X. Chen et al. Materials Science & Engineering A 688 (2017) 416–428

Table 1
Chemical composition of the ACDI in weight percent (wt%); Fe: balance.

Samples C Si Mn P S Cu Ni Cr Ti Mg Nb

1 3.87 1.64 0.30 0.04 0.02 0.45 0.57 0.02 0.03 0.05 0
2 3.86 1.67 0.32 0.03 0.02 0.41 0.70 0.02 0.03 0.04 0.04
3 3.92 1.69 0.25 0.03 0.02 0.44 0.64 0.03 0.03 0.04 0.06
4 3.91 1.76 0.33 0.06 0.02 0.46 0.60 0.02 0.03 0.03 0.08
5 3.86 1.72 0.30 0.04 0.02 0.42 0.71 0.03 0.03 0.04 0.11

Fig. 2. Optical micrographs showing the graphite morphology of the ACDIs with different Nb contents: (a) 0 wt%Nb, (b) 0.04 wt%Nb, (c) 0.06 wt%Nb, (d) 0.08 wt%Nb and (e) 0.11 wt%
Nb.

418
X. Chen et al. Materials Science & Engineering A 688 (2017) 416–428

If the shape of a particle is a perfect circle, the roundness value of

the graphite nodule is 1. The fewer bumps a particle have, the lower the
roundness value. The pearlite lamellar spacing was measured accord-
ing to the procedure reported by Voort et al. [36]. The procedure was
described as follows: selecting a circle on a high amplification micro-
graph containing pearlite, and counting the number of intersections
between the circumference and cementite layers. The pearlite lamellar
spacing was calculated by the following formula:
πD
S =
nM (4)
where s is the average pearlite lamellar spacing, D is the circle
diameter, n is the number of intersections between the circumference
and cementite layers, and M is the magnification of the selected
micrograph. In the present analysis, for each microstructure para-
meter, at least 40 circles were randomly selected from each specimen to
ensure an uncertainty less than 5%.
The detailed features of the microstructure were also characterized
Fig. 3. Nodularity, roundness of graphite nodules of the ACDIs with different Nb at high magnifications by Phenom ProX scanning electron microscope
contents. (SEM) in a backscattered electron (BSE) mode with a maximum
resolution of 10 nm in a backscattered mode, and maximum useful
magnification of 100,000. To maximize composition reading of the
Energy Dispersive Spectroscopy (EDS) data, an etchant was applied to
polished specimens for microscopic examination. Fractured surfaces of
tensile specimens were analyzed by the SEM in a secondary electrons
(SE) mode to ascertain the nature of fracture mechanisms. Thin
sections for transmission electron microscopy (TEM) were first ground
on a series of CARBIMET abrasive papers to thin foils with thickness of
~20 µm. The thin foils were further thinned by ion beam milling with
an incidence angle of 3–5° for about 1 h. The thin foils were examined
on a JEOL 2010 transmission electron microscope (TEM) equipped
with an energy dispersive X-ray spectrometer (EDX) at an operating
volt of 200 keV.

2.3. Tensile testing

Tensile specimens with a gage length of 0.06 m and a diameter of

Fig. 4. Graphite nodule count and average diameters of graphite nodules as a function of
Nb contents in the ACDIs.
block, an optical microscope (Axio Cam MRc5) was used to analyze
graphite morphology, and phase constitution including graphite nodule
conunts, ferrite and pearlite content in the matrix microstructure. The
nodularity, graphite nodule count and ferrite, pearlite content were
statistically determined by Image-Pro Plus software in accordance to
ASTM 247 [34]. Graphite particles were classified into five groups
based on the area percentage ratio calculated by the following equation
Ag
C=
Asc
where C is a constant for a specific graphite particle, Ag is the measured
area of a selected graphite particle, and Asc is the calculated area of a
minimized circle surrounding the selected graphite particle. The
constant ( C) value for each individual graphite particle was rounded
off to one decimal point in percentage. The nodularity ( N) of graphite
particles was calculated by
n
∑i =0 Ci ng
N= n
∑i =0 ng
where ng is the number of graphite particles in each classified group
with a specific constant of Ci.
The roundness is defined by the following equation [35]:
perimeter 2
roundness =
4 × π × area
0.01 m were prepared in accordance with ASTM E8 [37]. Tensile
testing was carried out on the 20-t SANS machine with a strain rate of
3×10−3 s−1 in room temperature. The test data was recorded by a data-
acquisition system which was connected to the tensile testing machine.
The tensile properties, including ultimate tensile strength (UTS), yield
strength (YS) and elongation (ef) obtained by the extraction of recorded
data based on the average of three tests.
3. Results
3.1. Microstructure
(1)
Fig. 2 presents optical micrographs showing the unetched matrices
of all the experimental ACDIs. It is evident that all the tested ACDIs
had a satisfactory level of nodularity. The distribution of graphite in the
matrices of the ACDIs was homogeneous. Fig. 3 shows the influence of
Nb addition on the nodularity and roundness of graphite of the ACDI.
The nodularity decreased from 92.4% to 84.5%, and the roundness
value decreased from 0.89 to 0.81 as Nb content increased from 0 to
0.11 wt%.
Fig. 4 displays the graphite nodule count and average diameter of
(2) graphite nodules as a function of Nb content in the ACDIs. It can be
seen from Fig. 4 that, with increasing Nb content from 0 to 0.11 wt%,
the graphite nodule count decreased from 287/mm2 to 212/mm2, and
the average diameter of graphite nodules increased from 17.2 to
20.2 µm.
Fig. 5 gives optical micrographs showing the phase constituents of
graphite, ferrite and pearlite in the microstructure of the etched ACDIs.
(3)
It is obvious from Fig. 5 that the ferrite rim around graphite nodules

419
X. Chen et al. Materials Science & Engineering A 688 (2017) 416–428

Fig. 5. Optical micrographs showing the phase constituents of graphite, ferrite and pearlite in microstructure of the etched ACDIs with different Nb contents: (a) 0 wt% Nb，(b) 0.04 wt
% Nb，(c) 0.06 wt% Nb，(d)0.08 wt% Nb and (e) 0.11 wt% Nb.

(“bullseye”) became notably thinner, and the pearlite content rose with
increasing niobium contents. Fig. 6 reveals the effect of Nb addition on
the pearlite and graphite volume percentages in the Nb-alloyed ACDIs.
Fig. 7 shows the pearlite microstructure in the tested ACDIs with
the different Nb contents by SEM micrographs in a backscattered
electron (BSE) mode. A selected circle in Fig. 7(d) represented the area
for the measurement of the pearlite lamellae spacing.
Fig. 8 presents the TEM results showing the nano and micro-sized
niobium carbides present in the ACDI with 0.08 wt% Nb addition. The
Nb-containing precipitates were present in the form of a single nano-
sized particle given in Fig. 8(a) or micro-sized particle clusters revealed
in Fig. 8(b). The probe crossed the nano-size particle along the solid
line in Fig. 8(c–f). When the probe went from the graphite to the
particle, the elements of Ti, Nb and Fe were detected. The selected area
diffraction pattern in Fig. 8(e) revealed the Nb-containing phase was
(Nb, Ti) C with a cubic structure. The detection of iron might result
from the inherent content of the ACDI.

420
X. Chen et al.
Fig. 6. Effect of Nb addition on the volume percentage of the pearlite and graphite in the
ACDIs.
3.2. Tensile properties and fracture
The representative engineering stress-strain curves of all the five
ACDIs measured at room temperature are given in Fig. 9. Table 2
summarizes the yield strength (YS), ultimate tensile strength (UTS)
and elongation (ef) values of the tested Nb-alloyed ACDIs in details,
which were extracted from the curves in Fig. 9. The small addition of
Nb at 0.04 wt% increased the YS, UTS and ef of the ACDI from 373,
671 MPa and 7.0% to 380, 681 MPa and 7.0–7.4% by a very minor rise
of 1.9%, 1.5% and 5.7%, respectively, as compared to Nb-free ACDI. As
Nb addition increased from 0.04 to 0.08 wt%, the YS, UTS and ef of the
Nb-alloyed ACDIs arose to 418 by 10%, 746 MPa by 9.5% and 8.0% by
8.1% from 380, 681 MPa and 7.4%, respectively. With increasing Nb
content from 0 to 0.08 wt%, both the YS, UTS and elongation of the
ACDIs increased from 373, 671 MPa and 7.0% to 418, 746 MPa and
8.0% by 12.1%, 11.2% and 14.3%. As Nb further rose from 0.08 to
0.11 wt%, however, the YS, UTS and Ef decreased slightly from 418,
746 MPa and 8.0% to 382, 690 MPa and 7.4% by 8.6%, 7.5% and 7.5%.
SEM fractographs in SE mode present the fracture surfaces of the
Nb-free ACDI and the ACDIs with the highest Nb content (0.08 wt%)
after tensile testing as shown in Fig. 10. The direct comparison between
the Nb-free and Nb-containing ACDIs revealed that there was almost
little difference in the features of their fracture surfaces.
4. Discussion
4.1. Microstructure analysis
The nodularity and roundness of graphite nodule are the major
microstructural criteria to evaluate graphite morphology in ductile
iron. High nodularity and low roundness values usually indicate that
the graphite nodule is close to the shape of an ideal sphere [35]. It is
well known that the nodularity is an important factor influencing the
tensile strength since the micro-cracking is more easily initiated at the
corners of irregular graphite nodules due to stress concentration [9].
The addition of Nb up to 0.11 wt% in the ACDI had a limited effect on
the nodularity and roundness of graphite nodule.
Table 3 lists the content of the phase constituents, pearlite and
graphite, in the matrix of the ACDIs with different Nb contents. The
improved roundness of graphite nodules reduced stress concentration
in the surrounding matrix. The addition of Nb led to an increase in the
pearlite content from 74.4% to 85.8% and a minor decrease in the
graphite content from 9.9% to 8.5% as compared to the Nb-free ACDI.
The optical microstructure observation suggested that the influence
of niobium alloying to ACDI with regard to the effect of graphite
421
Materials Science & Engineering A 688 (2017) 416–428
nodules could be considered negligible up to a level of 0.08 wt% Nb.
Additional Nb content over 0.08 wt% in the ACDI deteriorated the
morphology of graphite nodules to an extent, which could reduce its
mechanical properties (Fig. 4).
The properties of the pearlite phase depend on the detailed
microstructure characteristics such as the pearlite lamellae spacing.
The average values of pearlite lamellar spacing measured in the ACDI
with the different Nb contents are summarized in Table 4. The addition
of Nb to the ACDI gave rise to a decrease in the pearlite lamellar
spacing from 1.04 to 0.79 µm when the Nb content increased from 0 to
0.11 wt%. It has been demonstrated by the past research on pearlitic
irons and steels that the presence of a fine lamellar spacing led to high
strengths and especially in improved toughness [16–18,22].
It is evident that the addition of higher amounts of niobium to the
ACDIs resulted in two competing effects which were directly related to
microstructure development. The degradation of graphite nodule
morphology with increasing niobium contents caused a reduction in
mechanical properties. But, the refinement of the pearlite microstruc-
ture resulting from the increased usage of niobium improved mechan-
ical properties of the ACDIs.
4.2. Metallurgical consideration of niobium addition
It has been indicated by the previous work [15] that niobium
exhibits high affinity with carbon to form carbides and has low
solubility especially in iron alloys which contain relatively high carbon
content, such as cast iron. Depending on the actual concentration of
carbon and niobium in the carbon-containing alloy, four scenarios
might happen:
1. Primary precipitation of Nb-containing carbides in the melt;
2. Precipitation in the solid-liquid interval;
3. Precipitation in austenite; and
4. Precipitation in ferrite.
The first two scenarios are reflected in the phase diagram repre-
senting the appearance of niobium in eutectic iron with 4.23 wt% C
[26], which closely corresponds to the carbon equivalents of the ACDIs
employed in the present study. Above the melting temperature of
1500 °C, substantial amount of niobium could be dissolved in the
ACDIs. With decreasing temperature below the liquidus, the precipita-
tion of primary niobium carbide out of the liquid takes place due to the
reduced solubility of niobium. These primary precipitates have a cubic
morphology and are several micrometers in size. As can be seen from
the phase diagram given in Fig. 11, almost no Nb carbide is
precipitated from the liquid, and only austenite (γ-iron) and graphite
form in the liquid while Nb addition is low and less than 0.07 wt%.
The amount of solute niobium in austenite can be calculated from
the solubility product. Ohtani et al. [38] proposed a solubility product
that should be particularly applicable in high carbon iron alloys as
follows:
4880 ⎛ 1590 ⎞
log[Nbwt %]∙ log[Cwt %]= +1. 18 + [Cwt %]∙ ⎜ −0. 1⎟
T ⎝ T ⎠ (5)
where T is the absolute temperature. The carbon correction term in this
formula accounts for a marked reduction in the chemical potential for
Nb at high carbon concentration. Inserting these carbon concentrations
in Eq. (5), the maximum solubility of Nb as a function of temperature is
calculated and displayed in Fig. 12, which considers the temperature
range of 700–1200 °C since the three phase field (austenite, graphite
and liquid) ends up at a temperature lower than 1200 °C. At the onset
of austenite formation from the liquid, its soluble Nb content is around
0.1 wt%. The Nb solubility in austenite subsequently increases to reach
a maximum at the eutectic temperature. During equilibrium solidifica-
tion, the Nb solubility in austenite decreases significantly as the
X. Chen et al. Materials Science & Engineering A 688 (2017) 416–428

Fig. 7. SEM micrographs in BSE mode showing pearlite morphology in the ACDIs with different Nb contents: (a) 0 wt% Nb，(b) 0.04 wt% Nb，(c) 0.06 wt% Nb，(d)0.08 wt% Nb and
(e) 0.11 wt% Nb.

422
X. Chen et al. Materials Science & Engineering A 688 (2017) 416–428

Fig. 8. TEM micrographs of (a) a nano-sized (Nb, Ti)C particle, (b) micro-sized (Nb, Ti) C particle clusters embedded in the graphite nodules of the ACDI with 0.08 wt% Nb, (c) line
scans and the corresponding line scanning pattern for the nano-sized (Nb, Ti)C particle, EDS patterns for (d) Ti, (e) Nb and (f) Fe, and (g) diffraction pattern of nano-sized (Nb, Ti)C
phase.

423
X. Chen et al.
Fig. 9. Typical engineering stress-strain curves of the ACDIs with different Nb contents.
Table 2
Tensile properties, YS, UTS and elongation of the ACDIs varying with Nb content.
sample YS, MPa UTS, MPa
0% 373 671
0.04% 380 681
0.06% 386 715
0.08% 418 746
0.11% 382 690
temperature drops by hundreds of degrees. It is important to notice
that these values hold for equilibrium conditions. In conventional
solidification processes under non-equilibrium conditions such as
casting, fast cooling might suppress the precipitation of niobium
carbide especially in the solid phase. Consequently, a high amount of
Nb is expected to exist as remaining solutes in austenite especially
when no deformation leading to strain-induced precipitation is applied.
For casting processes, this condition should prevail.
Since the carbon equivalents of the ACDIs used in the present study
were close to the carbon percentage of the eutectic iron alloy given in
Fig. 11, their Nb solubility behaviors should be similar to each other. It
could be expected that no Nb precipitated in the liquid phase and a
substantial amount of Nb remained dissolved as solutes until the
pearlite formation temperature reached. It is reported that the
presence of solute Nb reduces the diffusivity of carbon and exert strong
solute drag on grain boundaries. Pinning by nano-size Nb-containing
precipitates as well as drag of solute Nb obstructed grain growth in
austenite [22]. The reduced carbon diffusivity by solute Nb delayed the
pearlite transformation and reduced lamellar spacing. On the other
hand, solute niobium hampered the carbon diffusion from the auste-
nitic periphery to the graphite nodules. Lacking substantial depletion of
carbon in the periphery, ferrite formation was impeded. These effects
should be reflected in the Nb-containing ACDIs employed in the
present study.
The observation on microstructural features in the presently
studied niobium alloyed ACDIs suggested that, the low niobium
addition might had a limited influence on the formation of graphite
nodules since most of the niobium remained dissolved. The solute
niobium could lead to the refinement of pearlite lamellar. With
increasing Nb addition over 0.08 wt%, the NbC precipitation in the
liquid could take place based on the phase diagram given in Fig. 11.
These particles had an influence on the graphite nodule morphology
and size.
Despite no indication of Ti in the NbC given in the Fe-Nb isopleth
Materials Science & Engineering A 688 (2017) 416–428
section (Fig. 11), the presence of Ti identified by TEM in the Nb-
containing precipitates should be attributed to the fact that Ti was
available as an impurity in the studied ACDIs, which has a high affinity
with carbon [39]. The identified (Nb, Ti)C particles were situated inside
the graphite nodules. This observation indicated that the (Nb, Ti)C
precipitates were the primary phase present in the liquid phase. They
formed before the austenite-graphite eutectic reaction during solidifi-
cation, which could efficiently act as nucleation sites for the formation
of graphite nodules. The size and distribution of the Nb-containing
particles influenced the size, morphology and number of graphite
nodules. Large Nb-containing particles or particle clusters promoted
larger graphite nodules with low roundness. Nano precipitates also had
an inoculating effect promoting small and round shaped graphite
nodules. The stability of (Ti, Nb)C in steels was indeed predicted by
Jang et al. [40] using first-principles calculations.
4.3. Tensile behavior
The results of tensile testing indicated that the Nb addition affected
the tensile and yield strengths and elongation of the ACDIs. The YS,
UTS and Ef of the four Nb-containing alloys were higher than those of
the Nb-free alloys. The moderate pearlite lamellar refinement (Fig. 7
and Table 4) and a small increase of pearlite volume percentage
(Table 3) should be responsible for the minor rise of the tensile
ef %
strength and elongation. In the range of 0.04 and 0.08 wt% Nb, the
7.0 increases in the YS by 10%, UTS by 9.5% and ef by 8.1% might result
7.4 from several competing effects. The ACDIs experienced the insignif-
7.8 icant deterioration of the graphite nodule morphology, and their
8.0
pearlite volume percentage increased, and their pearlite lamellar
7.4
spacing reduced. When Nb addition further increased to 0.11 from
0.08 wt%, the weakening effects on the tensile properties of the Nb-
alloyed ACDIs caused by the progressive degradation of the graphite
nodule morphology offset the contribution of the refined pearlite
lamellar structure. Hence, the highest tensile strength (746 MPa) and
elongation (8.0%) of the Nb-alloyed ACDIs was achieved at 0.08 wt%
Nb, which could be considered as the optimum value of tensile
properties. Overall, the determined tensile properties of the Nb-alloyed
ACDIs are fully in line with the discovered microstructural features of
the analyzed alloys discussed in the preceding section.
It is indicated by Fraś et al. [30] that the YS, UTS and ef of the
ductile iron alloyed by 0.038 wt% Nb were 272 MPa, 426 MPa and
23%, respectively. In the present work, the ductile iron containing
0.04 wt% Nb exhibited the YS, UTS and ef of 380 MPa, 681 MPa and
7.4%, respectively. The comparison with the results of Fraś et al. [30]
indicated the strength improvement achieved in the present study was
evidently significant. The appearance of the relatively low elongation of
7.4% should at least be attributed to the fact that the limited presence
of the ferrite was in the matrix structure of the 0.04 wt% Nb-alloyed
ductile iron compared with that revealed in Ref. [30].
4.4. Strain hardening
For the analysis of deformation behavior, true stress (σt) and strain
(εt) needs to be determined by the following equations
σt =σe (εe + 1) (6)
ε t = ln(εe + 1) (7)
where σt is the true stress, σe is the engineering stress, εt is the true
strain, and εe is the engineering strain.
Fig. 13 displays the typical true stress-strain curves of the ACDIs
with different Nb contents. They displayed similar trends with two
stages of elastic and plastic deformations. Upon the application of
tensile stresses, the ACDIs started experiencing elastic deformation.
When the yield point was passed, the ACDIs deformed plastically until
fracture.
424
X. Chen et al. Materials Science & Engineering A 688 (2017) 416–428

Fig. 10. SEM fractographs in SE mode showing the fracture surfaces of the ACDIs with the Nb contents: (a) low magnification and (b) high magnification for 0 wt%Nb; and (c) low
magnification and (d) high magnification for 0.08 wt% Nb.

Table 3 The plastic deformation of the stress-strain curve for metals can
Phase content in the matrix with different Nb contents in the ACDI. described by the power law:
Nb content (wt%) 0.0 0.04 0.06 0.08 0.11 σ = K εn (8)

Pearlite (vol%) 74.4 80.2 79.1 85.2 85.8 where K and n are empirical constants. The regression analysis
Graphite (vol%) 9.9 9.2 9.4 8.6 8.5 indicates that the power law was in good agreement with the tensile
results. The numerical values of these constants in the above equation
with the regression coefficients are listed in Table 5. The above
Table 4
Variation of pearlite lamellar spacing of the ACDI with the different Nb contents.
equation can be differentiated to obtain strain-hardening rates (dσ/dε).
All the tested ACDIs exhibited a pronounced strain hardening
Nb content (wt%) 0.0 0.04 0.06 0.08 0.11 behavior during plastic deformation. The strain hardening rates of
the Nb-alloyed ACDIs were determined based on the measurements of
Pearlite lamellar spacing (µm) 1.04 0.97 0.89 0.87 0.79
tensile testing. Fig. 14 displays the variation of the strain hardening
rates as a function of true strain for all the studied ACDIs with different
Nb contents. It is evident that, for all the five tested ACDIs, their strain-
hardening rates decreased with increasing true strains. As can be seen

425
X. Chen et al. Materials Science & Engineering A 688 (2017) 416–428

1500 Table 5
Fe + 4.23% C Best fit parameters of power equation.
1450
Sample K (MPa) n R2

1400 0% 1418 0.2521 0.9945

0.04% 1461.8 0.2557 0.9959
Liquid (L)
0.06% 1503.8 0.2567 0.9953
Temperature (°C)

1350 0.08% 1555.3 0.2489 0.9954

L + NbC 0.11% 1485.4 0.2554 0.9955
1300

from Fig. 14(a), the ACDI with 0.08 wt% had a high strain- hardening
1250 L+ +
graphite
rate (24,325 MPa) with respect to the Nb-free ACDI (21,190 MPa) at
L + NbC + + graphite
the onset of plastic deformation, which resulted in an increase of
1200 14.8%. For the Nb contents up to 0.08 wt%, the strain hardening rates
of the Nb-containing ACDIs increased with increasing Nb addition, and
1150 were consistently higher over the entire plastic deformation stage of the
+ NbC + graphite tensile strain range than that of the Nb-free ACDI. As depicted in
1100 Fig. 13(b), an additional introdution in the Nb content from 0.08 to
0 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.11 wt% led to a minor decrease in the strain hardening rate of the
Niobium content (wt%) Nb-alloyed ACDI by only 5.3%. Hence, the content of 0.08 wt% Nb
Fig. 11. Phases diagram of Nb in eutectic iron (Fe+4.23 wt%C) [25]. could be considered as an optimum level in terms of tensile behavior of
the Nb-alloyed ACDIs. The observation on the strain hardening
behavior suggested that the Nb addition enhanced the ACDI capability
of spontaneously strengthening itself increasingly to a large extent, in
response to a slight plastic deformation after yielding.

4.5. Fracture behavior

The facture behavior of material plays a significant role with regard

Fig. 12. Maximum solubility of niobium in austenite under equilibrium conditions.
to safety and reliability during service. In the presented two cases,
typical quasi-cleavage fracture was observed. The dimple-type features
had a loose connection to the plastic deformation of the pearlite, but
rather related directly to extracted graphite nodules. Tear ridges
observed on the cleavage plains were formed during linking up of
microcracks with the advancing fracture front. The favorable round-
ness of the graphite nodules reduced the stress concentration at the
matrix/graphite interface, which delayed the onset of cracking. In
addition, the refinement of both the prior-austenite grains and the
pearlite lamellar structure provided barriers to crack propagation,
which were beneficial to extended plastic deformation under high
tensile loading. Considering the factors affecting the fracture modes,
Nb addition in the range of 0.06–0.08 wt% should be beneficial
sufficiently to the engineering performance of the ACDI. The similar
fracture modes for the unalloyed and Ni or Co-alloyed ductile irons
with the carbon equivalent of 4.4% were also observed in the work by
Hsu et al. [18].

5. Conclusions

The influence of niobium addition on the microstructure and tensile

properties of as-cast ductile iron has been investigated. The congruent
relationship between microstructure, constitution of niobium as solute
or precipitate as well as mechanical properties is identified. Concerning
the effects of niobium alloying, the following conclusions can be drawn:

1. The niobium additions up to 0.08 wt% into the base ductile iron had
insignificant influence on the graphite nodule morphology and
nodule count. In these ACDI alloys, niobium remained in solution
before and after solidification. Increasing amount of solute niobium
promoted pearlite formation and refines lamellar spacing.
2. The relatively high amount of niobium addition, particularly at a
level of 0.11 wt% resulted in partial precipitation of Nb carbide
Fig. 13. Typical true stress-strain curves of the ACDIs with different Nb contents.
particles in the liquid phase. The precipitated particles could provide
nucleation sites for graphite nodules. But, the larger primary Nb
carbide particles as well as particle clusters might exhibit a detri-

426
X. Chen et al.
Fig. 14. Strain hardening of the ACDIs with the Nb contents of (a) 0–0.08 wt%, and (b) 0.08 wt% and 0.11 wt%.
mental effect on the graphite nodule morphology and size. The
selected area diffraction pattern of TEM analyses revealed the Nb-
containing phase was (Nb, Ti) C with a cubic structure.
3. The optimum niobium addition to the ACDI was found to be around
0.08 wt%, which resulted in the highest tensile strength of 746 MPa
and elongation of 8.0% among the Nb-containing ACDIs. With the
optimal microstructural constitution, the tensile strength and elon-
gation have an improvement of 11.2% and 14.3%, respectively,
compared to the niobium-free ductile iron. The significant micro-
structural enhancement of the pearlite phase over the deterioration
of the graphite morphology in the ACDI by the Nb addition should
be responsible for the significant increase in the tensile properties.
4. The analysis of plastic deformation indicated that the strain hard-
ening rates of the Nb-containing ACDIs increased with increasing
Nb addition, and were consistently higher over the entire plastic
deformation stage of the tensile strain range than that of the Nb-free
ACDI, which enhanced the ACDI capability of spontaneously
strengthening itself increasingly to a large extent, in response to a
slight plastic deformation after yielding.
Acknowledgement
The authors heartily acknowledge the financial support from CITIC-
CBMM Nb Steel Research and Development Fund (No. 2013-D069)
and Natural Science Foundation of China (No. 51404150). One of the
co-author, Henry Hu, would like to thank the Natural Sciences and
Engineering Research Council of Canada (NSERC) and the University
of Windsor for supporting this work.
References
[1] M.F. Ashby, D.R.H. Jones, Engineering Materials 2: An Introduction to
Microstructures, Procssing and Design, Fourth edition, Butterworth Heinemann,
London, 2013, pp. 3–12.
[2] John Campbell, Casting, Second edition, Butterworth-Heinemann, London, 2003,
pp. 70–231.
[3] J.R. Davis, Cast Irons, ASM Specialty Handbook, ASM International, Materials
Park, OH, USA, 1996, pp. 32–79.
[4] H. Morrogh, W.J. Williams, The production of nodular graphite structures in cast
iron, J. Iron Steel Inst. Lond. 158 (1948) 306–322.
[5] C. Labrecque, M. Gagné, Review ductile iron: fifty years of continuous develop-
ment, Can. Metall. Q. 37 (5) (1998) 343–378.
[6] A. Meena, M. El Mansori, Study of dry and minimum quantity lubrication drilling
of novel austempered ductile iron (ADI) for automotive applications, Wear 271
(2011) 2412–2416.
[7] S. Schoenborn, H. Kaufmann, C.M. Sonsin, R. Heim, Cumulative damage of high-
strength cast iron alloys for automotive applications, Procedia Eng. 101 (2015)
440–4449.
[8] Q. Cai, B. Wei, Recent development of ductile cast iron production technology in
427
Materials Science & Engineering A 688 (2017) 416–428
China, China Foundry 2 (2008) 82–91.
[9] F. Iacoviello, O. Di Bartolomeo, V. Di Cocco, V. Piacente, Damage micromechan-
isms in ferritic-pearlitic ductile cast iron, J. Mater. Sci. Eng. A 478 (2008) 181–186.
[10] J.L. Hernández-Rivera, R.E. Campos Cambranis, A. de la Garza, Study of micro-
structural evolution and mechanical properties exhibited by nonalloyed ductile iron
during conventional and stepped austempering heat treatment, Mater. Des. 32
(2011) 4756–4762.
[11] A.R. Kiani-Rashid, Influence of austenitising conditions and aluminum content on
microstructure and properties of ductile irons, J. Alloy. Compd. 470 (2009)
323–327.
[12] Y. Sun, S. Hu, Z. Xiao, S. You, J. Zhao, Y. Lv, Effects of nickel on low-temperature
impact toughness and corrosion resistance of high-ductility ductile iron, Mater.
Des. 41 (2012) 37–42.
[13] G.S. Cho, K.H. Choe, K.W. Lee, A. Ikenaga, Effects of alloying elements on the
microstructures and mechanical properties of heavy section ductile cast iron, J.
Mater. Sci. Technol. 23 (2007) 97–101.
[14] P.W. Shelton, A.A. Bonner, The effect of copper additions to the mechanical
properties of austempered ductile iron (ADI), J. Mater. Process. Technol. 173
(2006) 269–274.
[15] A.S. Benam, Effect of alloying elements on austempered ductile iron (ADI)
properties and its process: review, China Foundry 121 (17) (2015) 1672–6421.
[16] G. Xu, X. Gan, The development of Ti-alloyed high strength microalloy steel, Mater.
Des. 31 (2010) 2891–2896.
[17] F. Katsuki, K. Watari, Abrasive wear behavior of a pearlitic (0.4%C) steel
microalloyed with vanadium, Wear 264 (2008) 331–336.
[18] Cheng-Hsun Hsu, Ming-Li Chen, Chia-Jung Hu, Microstructure and mechanical
properties of 4% cobalt and nickel alloyed ductile irons, Mater. Sci. Eng. A 444
(2007) 339–346.
[19] N. Matougui a,b, D. Piot a,n, M.L. Fares b, F. Montheillet a, S.L. Semiatin, Influence
of niobium solutes on the mechanical behavior of nickel during hot working, Mater.
Sci. Eng. A 586 (2013) 350–357.
[20] N.D. Evans, P.J. Maziasz, R.W. Swindeman, G.D. Smith, Microstructure and phase
stability in INCONEL alloy 740 during creep, Scr. Mater. 51 (2004) 503–507.
[21] G.D. Smith, S.J. Patel, The role of niobium in wrought precipitation-hardened
nickel-based alloys, Superalloys 718, 625, 706 and Derivatives 2005, TMS, 2005,
pp.135–154.
[22] W.B. Morrison, Microalloy steels – the beginning, Mater. Sci. Technol. 25 (9)
(2009) 1066–1073.
[23] H. Mohrbacher, Niobium alloyed high strength steels for automotive applications,
Int. J. ISSI 3 (1) (2006) 13–19.
[24] A.I. Filhoa, R.V. Silvaa, W.S. Cardosoa, L.C. Casteletti, Effect of niobium in the
phase transformation and corrosion resistance of one austenitic-ferritic stainless
steel, Mater. Res. 17 (4) (2014) 801–806.
[25] Zhou Wenbin, Zhu Hongbo, Zheng Dengke, Zheng Hongxing, Hua Qin, Zhai Qijie,
Nb alloying effect in high carbon equivalent grey cast iron, China Foundry 1 (8)
(2011) 36–40.
[26] H. Mohrbacher, K. Hulka, Niobium Information: Niobium in cast iron, CBMM
technical brochure, published in 2004.
[27] A.O. Devecili, R. Yakut, The effect of Nb supplement on material characteristics of
iron with lamellar graphite, Adv. Mater. Sci. Eng. 2014 (2014) 1–5.
[28] H. Mohrbacher, Q. Zhai, Niobium alloying in grey cast iron for vehicle brake discs,
2011. in: TMS conf. Proceedings Materials Science and Technology (MS & T), San
Diego, California, the USA, 2011, pp. 434–445.
[29] Zhu Hong-bo, Yan Yong-sheng, Sun Xiao-liang, Hua Qin, Zhai Qi-jie, Effects of Nb
on high temperature tensile strength and oxidation resistance of gray cast iron,
Mod. Cast. Iron 2 (2011) 49–51.
[30] E. Fraś, M. Górny, M. Kawalec, Effect of small additions of vanadium and niobium
on structure and mechanical properties of ductile iron, Arch. Foundry Eng. 7 (1)
(2007) 89–92.
X. Chen et al.
[31] T.N.F. Souza, R.A.P.S. Nogueirac, F.J.S. Francob, M.T.P. Aguilard, P.R. Cetlin,
Mechanical and microstructural characterization of nodular cast iron (NCI) with
niobium additions, Mater. Res. 17 (5) (2014) 1167–1172.
[32] S.K. Alias, B. Abdullah, A. Jaffar, A.H. Abdullah, N. Jenal, Development of high
strength ductile iron with niobium addition, Adv. Mater. Res. 576 (2012) 366.
[33] A. Bedolla-Jacuinde, E. Solis, B. Hernandez, Effect of niobium in medium alloyed
ductile cast irons, Int. J. Cast. Met. Res. 16 (5) (2003) 1–6.
[34] ASTM 247: Annual Book of ASTM Standards, vol.01. 02, 2012, pp. 117–118.
[35] Xu Weixing, Jin Zhihua, Gu Yan, Numerical processing of graphite nodularity [J],
Phys. Chem. Test. 45 (2009) 544–545.
[36] G.F. Vander Voort, A. Roosz, Measurement of the interlamellar spacing of pearlite,
428
Materials Science & Engineering A 688 (2017) 416–428
Metallography 17 (1984) 1–17.
[37] ASTM E8M: Annual Book of ASTM Standards, vol.03. 01, 2012, pp. 77–97.
[38] H. Ohtani, T. Tanaka, M. Hasebe, T. Nishizawa, Solubility of NaCl-type carbides
(NbC, VC and TiC) in Austenite, in: Proc. of the Japan-Canada Seminar, Tokyo,
Japan. Dec. (1985) J-7-1 – J-7-12.
[39] V. Kuzucu, M. Aksoy, M.H. Korkut, The effect of strong carbide-forming elements
such as Mo, Ti, V and Nb on the microstructure of ferritic stainless steel, J. Mater.
Process. Technol. 82 (1998) 165–171.
[40] Jae Hoon Jang, Chang-Hoon Lee, Yoon-Uk Heo, Dong-Woo Suh, Stability of (Ti,M)
C (M = Nb, V, Mo and W) carbide in steels using first-principles calculations, Acta
Mater. 60 (2012) 208–217.